Advanced Efficient Feature Selection Integrating Augmented Extreme Learning Machine and Particle Swarm Optimization for Predicting Nitrogen Use Efficiency and Yield in Corn
, Isa Ebtehaj 1, Gabriel Deslauriers 2, Alain N. Rousseau 3, Hossein Bonakdari 4 and Jacynthe Dessureault-Rompré 1Round 1
Reviewer 1 Report
Comments and Suggestions for Authors
The study is complex and interesting, as it includes 84 sites for data collection. Although it is well done, I expected more information on the agronomic side, as it was submitted to Agronomy.
At Introduction we find a lot of information about corn cultivation and the amount of fertilizer applied, and the results are exclusively on Learning Machine Models. From this point of view the paper is unbalanced, the Introduction should contain more bibliographical study related to ML, and the Results part should provide more data related to soil health indicators and nitrogen fertilization.
In Material and Methods there is not enough information about the experimental design. For example, different soil tillage can influence the evolution of soil nitrogen (Conventional Plowing, Reduced Tillage or Direct Drilling).
All applied formulas must be accompanied by the bibliographical source (as long as they do not belong to the authors of the work).
Please correct figure 8 in the Results part and try to present the results in comparison with other studies carried out and especially to emphasize their applicability in a concrete way.
The conclusions are not an answer to the research hypothesis, but rather represent new hypotheses for future research.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors
The paper introduces a method combining Augmented Extreme Learning Machine (AELM) and Particle Swarm Optimization (PSO) for feature selection and predictive modeling of nitrogen use efficiency (NUE) and yield in corn. While the study is well-structured and provides detailed methodological explanations, the core methodology closely aligns with existing approaches, raising concerns about its novelty:
We recommend that :
- the authors consider sharing the source code to facilitate replication, further research, and potential improvements by the academic community.
- Further experiments or additional data analyses are necessary to enhance the research's robustness and address any remaining gaps.
- Enhance the visual presentation of workflows and figures for better comprehension: Exemple, improve the quality of the figure 2 by redesigning it as a workflow diagram to better illustrate the proposed methodology.
- Explain how this manuscript advances this field of research and/or contributes something new to the literature.
Provide a more detailed explanation of how PSO parameters (e.g., inertia weight, learning coefficients) were determined.
- Add a table summarizing the performance metrics for all models across outputs for easier comparison.
Comments on the Quality of English Language
- Simplify overly complex sentences to improve flow.
- Ensure consistency in technical terminology (e.g., feature selection, optimization methods).
- Pay special attention to passages generated by AI, ensuring they meet academic writing expectations and do not compromise the manuscript's quality.
- Review for minor typographical or grammatical errors.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for Authors
The article has been improved, taking into account my initial suggestions.
Reviewer 2 Report
Comments and Suggestions for Authors
-
